Using grammars for pattern recognition in images: A systematic review
ACM Computing Surveys (CSUR)
| Hi-index | 0.00 |
An image can be stored as a graph which requires very less memory for the storage. A graph is pictorial representation of a finite relation between the entities with help of nodes and edges. Edges connect the nodes whenever the nodes are related. For converting an image into a graph, it is first converted to a binary format. Binary image is converted to a skeleton form which preserves the shape details efficiently. Skeleton is then converted to a shock graph which has structure like a tree. The hierarchy of nodes in the graph structure is decided by a Shock Graph Grammar. Binary images with different shapes have different skeletons and different tree structure. Number of features of the graph can be extracted which facilitate comparison of shapes using these features. Comparison of shapes using their Shock graphs provides a very effective way of object recognition. A graph can be divided into sub graphs. An object recognition frame work by comparing the sub graphs has been presented here.